What each procedure actually does to the knee

The two procedures share the same opening step. In standard microfracture, the surgeon makes a series of small perforations — typically 3–4 mm apart — through the base of the cartilage defect into the subchondral bone beneath. This releases bone marrow fluid rich in mesenchymal stem cells (MSCs), which pool across the defect and clot. Those MSCs carry the potential to differentiate into repair tissue and resurface the damaged area.

AMIC takes that identical step and adds one further action: a bi-layer type I/III collagen membrane — Chondro-Gide® is the most extensively studied device — is pressed and fixed over the microfractured bed, encasing the marrow clot within a three-dimensional scaffold. That is the entirety of the operative difference. There is no separate biopsy session, no laboratory cell-culture phase, and no second operation. Both procedures are completed in a single stage.

Historically, microfracture was the default surgical option for focal chondral defects under roughly 2 cm², chosen largely for its simplicity, low cost, and all-arthroscopic delivery. The scaffold addition in AMIC does not alter those practical advantages — but it does change what happens to the clot after the surgeon closes the knee. Understanding why the membrane matters requires looking at what an unsupported marrow clot does, and does not do, as it matures inside a loaded joint.

Why an unsupported marrow clot breaks down over time

The marrow clot that forms after microfracture sits in a mechanically hostile environment from the moment the surgeon closes the knee. Early joint movement generates shear forces capable of disrupting the unsupported haematoma before it can consolidate into stable repair tissue. When the clot does survive and mature, the tissue that forms is predominantly fibrocartilage — type I collagen-dominant — rather than the native articular cartilage it is meant to replace.

Native articular cartilage is type II collagen-dominant, with a highly organised matrix that gives it stiffness and resilience under repetitive loading. Fibrocartilage lacks that structural organisation: it is softer, less stiff, and wears faster under cyclic compression. This explains why clinical decline after microfracture tends to emerge between 18 and 36 months post-operatively as the fibrous repair tissue begins to break down under normal use. Kreuz et al. documented score deterioration within exactly that window; Solheim et al. reported fewer than 60% of patients maintaining satisfactory outcomes at three years.

The AMIC collagen membrane addresses both failure mechanisms. Fixed over the microfractured bed immediately after the perforations are made, it physically anchors the marrow clot and shields it from shear stress during early mobilisation. The scaffold's three-dimensional structure also provides physical guidance that supports MSC retention and chondrogenic differentiation toward a more hyaline-like — though not equivalent to native — repair phenotype.

The tissue-quality difference is measurable on imaging. In a pooled analysis of 29 studies with a minimum two-year follow-up, MOCART structural scores averaged 69.3 for AMIC versus 41.0 for standard microfracture (P=0.005), and adequate defect filling on MRI was recorded in 77.3% of AMIC cases compared with 47.9% after microfracture (P=0.008). These figures reflect what happens when the marrow clot is given structural support rather than left to consolidate on its own.

What a decade of follow-up data actually shows

Deciding between two surgical options based on the first two years of data would, in the case of AMIC and microfracture, give the wrong answer. Both procedures produce comparable improvement in the Modified Cincinnati Knee Score and VAS pain through the first 12 to 24 months — a pattern the only randomised controlled trial to have run for ten years makes explicit.

That trial, NCT02993510 (Volz 2024, n=47: 13 microfracture, 17 sutured AMIC, 17 glued AMIC), tracks what happens after that early convergence. Beyond year two, the microfracture group deteriorates progressively and significantly on both outcome measures; neither AMIC group shows the same pattern. Both sutured and glued AMIC arms remain clinically stable through the full decade of follow-up. At ten years, responder rates stand at approximately 83–88% for AMIC versus roughly 22% for standard microfracture — a gap that would be invisible to any study that stopped at the two-year mark.

Those figures carry particular weight because surgeons and patients routinely judge surgical success at the 12 or 24-month review — exactly the window in which the two procedures appear equivalent. The divergence only becomes visible with sustained follow-up, which is why shorter comparison studies have historically understated the difference between them.

Functional scoring data from the same 29-study meta-analysis discussed in the previous section extend that picture. IKDC subjective improvement was significantly greater for AMIC (45.9 versus 27.2, P<0.001). Lysholm and Tegner differences, however, were not statistically significant in the pooled analysis — a nuance worth stating plainly, though it does not reverse the overall direction of the evidence.

For a patient weighing these two procedures, the practical implication is that early equivalence is not a reliable proxy for durability: the difference between an 83% and a 22% responder rate at ten years is precisely what sustained follow-up data exist to reveal.

Which patients and defects are suitable for AMIC

Defect size is the most important practical filter. Standard microfracture is generally indicated for focal chondral lesions smaller than 2 cm²; beyond that threshold the unsupported clot becomes less reliable, and AMIC takes over. A retrospective series demonstrated sustained pain relief and functional gain in full-thickness defects larger than 2 cm² at seven-year follow-up, and the 2024 systematic review of 18 studies (490 patients) reported a mean treated defect of 3.47 cm² — confirming that the typical AMIC patient in clinical practice has a lesion well above what microfracture can sensibly address. Single-stage treatment has been reported for defects up to 4.5 cm² and beyond.

Age modifies the imaging picture but not the core clinical benefit. In an age-stratified cohort, patients under 45 showed a significant correlation between MOCART structural scores and IKDC functional scores; those over 45 still achieved meaningful improvement in pain and function, even where the MRI-to-clinical correlation was weaker. The technique appears safe and reproducible across the adult age range when other selection criteria are met.

Lesion geometry is a practical secondary consideration. The standard bilayer membrane suits flat or gently curved surfaces; a chitosan-based injectable scaffold variant (BST-CarGel) fills vertical, inverted, and irregularly shaped defects that a flat membrane cannot conform to. A prospective study covering 31 such defects in 21 patients reported Lysholm and KOOS improvements of 25.8 and 22.5 respectively at a mean follow-up of 42.5 months, with no complications or reoperations.

Two factors place a patient outside the indication. AMIC depends on a viable bone-marrow contribution from the microfracture perforations, so significant subchondral bone damage undermines that mechanism from the outset. Diffuse osteoarthritic change across the joint compartment is equally out of scope — the technique addresses focal defects, not widespread articular loss, and patients with advanced OA are better served by a different pathway. Knee alignment is also a meaningful background variable: loading asymmetry affects any cartilage repair, and a consultant assessment will routinely include that consideration alongside defect geometry.

Where AMIC fits among other cartilage repair options

AMIC occupies a well-defined middle position on the cartilage repair spectrum — more structured and durable than standalone microfracture, but less resource-intensive than two-stage cell-based procedures such as ACI or MACI, which require a biopsy, laboratory cell culture, and a second operation.

At up to two years post-surgery, outcomes with AMIC are broadly comparable to those achieved with MACI and arthroscopic minced cartilage implantation; pain and function improve across all three without significant difference between them. Structural MRI tells a more differentiated story at five years, when microfracture fares significantly worse in defect fill and tissue quality, while AMIC and cell-based approaches track more closely at that timeframe.

For defects larger than approximately 4–6 cm², particularly where posttraumatic bone loss or failed prior cartilage surgery is involved, osteochondral allograft (OCA) is typically the more appropriate choice — AMIC addresses the cartilage surface only and cannot restore damaged subchondral bone stock.

Moving to the other end of the size range, patients with focal defects under 2 cm² who prefer a non-surgical route have a further option worth discussing: the ChondroFiller injection, an ultrasound-guided outpatient treatment using an injectable collagen scaffold. This sits below AMIC on the intervention scale, is suited to a different patient profile, and should not be conflated with membrane-based techniques applied in a theatre setting.

One evidence gap is worth stating plainly. Studies have not yet established whether the structural improvements seen with AMIC on MRI translate into reduced rates of knee replacement at population scale. That is an honest limit of current follow-up data, not a reason to avoid the procedure — and it sets realistic expectations for what can and cannot yet be confirmed.

Getting an accurate assessment at Lincolnshire Knee

Selecting the right cartilage repair strategy depends on structural detail that symptoms alone cannot supply. Defect grade, size in cm², subchondral bone condition, and lower-limb alignment each influence whether microfracture, AMIC, or a different approach is appropriate — and that detail requires objective imaging and functional assessment, not history alone.

Lincolnshire Knee, part of the MSK Doctors group and based at Sleaford NG34 and Grantham NG31, offers a consultant-led pathway built around this level of precision. AI-assisted MRI analysis through onMRI™ provides cartilage segmentation, meniscus mapping, and T2 signal data directly relevant to procedure selection. MAI Motion® objective gait assessment adds a functional layer — how the knee is loading through movement — alongside the structural picture.

No GP referral is needed to access this pathway. To book a cartilage assessment, visit lincolnshireknee.co.uk.

1. [1] Microfracture and Microfracture Plus of the Knee Joint. (2024). https://doi.org/10.1016/j.csm.2024.10.003 https://doi.org/10.1016/j.csm.2024.10.003
2. [2] Enhancing Cartilage Repair: Surgical Approaches, Orthobiologics, and the Promise of Exosomes. (2024). https://doi.org/10.3390/life14091149 https://doi.org/10.3390/life14091149
3. [3] A randomized controlled trial demonstrating sustained benefit of AMIC® over microfracture: 10-year follow-up. (2024). https://doi.org/10.1007/s00590-024-03948-0 https://doi.org/10.1007/s00590-024-03948-0
4. [4] A systematic review on Autologous Matrix Induced Chondrogenesis (AMIC) for chondral knee defects. (2024). https://doi.org/10.1016/j.knee.2024.08.003 https://doi.org/10.1016/j.knee.2024.08.003
5. [5] Clinical and Radiological Outcomes After AMIC Versus Microfracture of the Knee: Systematic Review and Meta-analysis With Minimum 2-Year Follow-up. (2020). https://doi.org/10.1177/2325967120959280 https://doi.org/10.1177/2325967120959280
6. [6] AMIC for Focal Chondral Lesions of the Knee: 2-Year Follow-Up of Clinical, Proprioceptive, and Isokinetic Evaluation. (2022). https://doi.org/10.3390/jfb13040277 https://doi.org/10.3390/jfb13040277
7. [7] Injectable Scaffold with Microfracture using AMIC Technique: A Prospective Cohort Study. (2022). https://doi.org/10.5704/MOJ.2211.014 https://doi.org/10.5704/MOJ.2211.014
8. [8] Microfracture- and Xeno-Matrix-Induced Chondrogenesis for Knee: Age-Based Mid-Term Results. (2023). https://doi.org/10.3390/healthcare11222995 https://doi.org/10.3390/healthcare11222995